JPS63781A - Kanji input device - Google Patents

Abstract

PURPOSE:To stably enable a foreigner, or a child who has no knowledge for the stroke of a KANJI (Chinese character) to input the KANJI, by providing plural vertical, horizontal and oblique rule marks respectively in an input area, and inputting the KANJI as the aggregate of line segments on the rule mark. CONSTITUTION:Plural vertical, horizontal, and oblique (45 deg.) rule marks are drawn in a KANJI input area and an input is performed on the rule mark. An input data is inputted as the line segment on the rule mark, and next, the stroke is identified. In other words, the line segment continuing on the same rule mark is set as one line segment. And the start point of each obtained line segment is set at an upper and a left part. Furthermore, two line segments in which the end point and the start point of each obtained line segment coincide, are set as one stroke, and an order is given on the stroke starting from the one whose start point is positioned at the upper and left part. A result in which similar stroke identification is performed on each KANJI is preserved as a dictionary, and it is collated with the result in which the stroke identification of the input data is performed, thereby, an inputted KANJI can be recognized.

Description

[Detailed Description of the Invention] [Summary] The present invention provides a kanji input device used as an input section for electronic dictionaries, Japanese word processors, etc.
This is a unique system that does not depend on the strokes of the kanji by providing ruled lines in the input area and inputting the kanji along the ruled lines, so that even people and children who are unable to enter kanji while being aware of the strokes can input kanji. The strokes of the input kanji are identified, and the input kanji is multiplied by 9'fJ m by comparing the strokes with a dictionary created by similarly identifying the strokes.

[Industrial application field]

The present invention relates to electronic dictionaries, Japanese word processors, etc. that utilize an online handwritten character recognition function, and particularly relates to a kanji input method.

In such electronic dictionaries and Japanese word processors, kanji are input as a sequence of coordinate points from an input device such as a tablet. At this time, in order to deal with the fluctuation in the number of strokes that occurs when the - part is written in succession, a method has been taken in which the data when written in continuous characters is registered in a dictionary, but there is a limit due to the dictionary size. Therefore, it has been desired to develop a stable manual method.

In particular, electronic dictionaries are basically used by learners who have no knowledge of how to write kanji or the connections between strokes, etc., inputting kanji and looking up their meanings. Development is desired.

[Conventional technology]

The configuration of a conventional kanji human-powered device is shown in Figure 2. Conventional kanji input devices include an input unit 21 such as a tablet that inputs kanji characters as a sequence of coordinate points, a dictionary 23 that describes each kanji character as a sequence of coordinate points in units of strokes, and a system that matches the input with the dictionary. It is composed of a recognition unit 22 that performs recognition.

[Problem to be solved by the invention] Conventional kanji input devices receive input data in units of strokes based on the strokes of kanji, so foreigners and children who do not have knowledge of kanji strokes cannot - Because the strokes were divided into multiple strokes and written manually, or multiple strokes were manually written in one stroke, recognition became difficult and it was not possible to reliably write kanji manually.

[Means to solve the problem]

Draw multiple vertical, horizontal, and diagonal (45°) ruled lines in the kanji input area, and enter data in line 2 on the ruled lines (Figure 3).

Input data is input as line segments on ruled lines, and then strokes are identified using the following procedure.

■ Continuous line segments on the same ruled line are considered as one line segment.

■ For each line segment obtained in ■, set the starting point at the top/left (Figure 4).

■ For each line segment obtained in (■), one stroke is defined as two line segments in which the end point and the start point are defeated.

■ Order strokes starting from the top/left starting point (
Figure 5).

The results of similar stroke identification for each kanji are saved as a dictionary, and compared with the stroke identification results of the input data, the manually written kanji are recognized.

[Effect]

With the present invention, it is possible to input kanji without being tied to the strokes, so if the learner writes the constituent elements of the strokes while looking at the kanji shown on printed matter using an electronic dictionary, the kanji can be input. can do.

〔Example〕

FIG. 1 shows a configuration diagram of an embodiment according to the present invention.

This device includes an input section 11, a stroke identification section 12, a recognition section 13, and a dictionary 14.

The input unit 11 has a kanji manual area shown in FIG. 3, and the dictionary 14 stores the results of stroke identification processing according to the present invention for each kanji.

The recognition unit 13 matches and recognizes stroke units as before.

〔effect〕

According to the present invention, even foreigners and children who have no knowledge of the strokes of kanji can stably input kanji.

[Brief explanation of drawings]

Fig. 1 is a block diagram of a kanji input device according to an embodiment of the present invention, Fig. 2 is a block diagram of a conventional kanji input device, Fig. 3 is a diagram showing the kanji human power area, and Fig. 4 is a diagram showing the allocation of start and end points. FIG. 5 is a diagram showing the ordering of strokes. In FIG. 1, 11 is an input section, 12 is a stroke identification section, 13 is a recognition section, and 14 is a dictionary. Is it the lower of the 1 code code? 4 rows of Mitsubishi (5; 3 Figure 0: Beginning P Δ; End, . Starting place, ・y−1 more than ke 1 ri ii suzu 2 de ni d 2 ] ne 4
mouth

Claims (3)

[Claims] (1) In a kanji input device that is equipped with an input section that inputs kanji characters as a sequence of coordinate points, and a recognition section that matches and recognizes kanji characters stroke by stroke, vertical, horizontal, and diagonal ruled lines are added to the input area of the input section. A kanji input device characterized in that a plurality of kanji characters are provided, and kanji characters are input as a collection of line segments on the ruled lines. (2) The kanji input device according to claim 1, further comprising means for identifying a unique stroke that does not depend on strokes of a kanji from a set of line segments on the plurality of ruled lines. (3) Having a dictionary that describes each kanji as a set of the above-mentioned unique strokes, and by comparing the stroke identification result of the input data by the above-mentioned stroke identification means with the contents of the dictionary, the input kanji can be recognized. The kanji input device according to claim (2), characterized in that: